Why Do Some People Spend Money on the Product Features That They May Never Use?

ABSTRACT - Marketers often introduce products with new, additional features in an effort to differentiate them from other products in the market. Sometimes those features are well accepted and used by consumers; other times, they are purchased but not actually used at all. This paper addresses and deals with an interesting questio, why do some consumers tend to buy the product with additional features that they may never use? Study 1 shows that such tendency does exist among consumers in the marketplace. Study 2 explores some situational factors that bring about this phenomenon. It demonstrates that consumers who have future expectations about the use of the additional features, or those who are concerned about potential purchase failure are likely to buy the product with those features. Research implications of the study results are discussed.


Jaebeom Suh, Yong-Soon Kang, and Moonkyu Lee (1998) ,"Why Do Some People Spend Money on the Product Features That They May Never Use?", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 538-543.

Advances in Consumer Research Volume 25, 1998      Pages 538-543


Jaebeom Suh, University of Alabama at Tuscaloosa

Yong-Soon Kang, State University of New York at Binghamton

Moonkyu Lee, Yonsei University


Marketers often introduce products with new, additional features in an effort to differentiate them from other products in the market. Sometimes those features are well accepted and used by consumers; other times, they are purchased but not actually used at all. This paper addresses and deals with an interesting questio, why do some consumers tend to buy the product with additional features that they may never use? Study 1 shows that such tendency does exist among consumers in the marketplace. Study 2 explores some situational factors that bring about this phenomenon. It demonstrates that consumers who have future expectations about the use of the additional features, or those who are concerned about potential purchase failure are likely to buy the product with those features. Research implications of the study results are discussed.


Since launching a totally new product in the market can be risky because of the high costs and the low probability of success, marketers often "extend" their current products by adding a few features [Although features can be understood as a broader concept than functions, they will be used interchangeably here.] to those products. In fact, one can easily find such products with additional features in the marketplace, e.g., a camcorder with a Steady Shooting function, a stereo system with Woofer Sound, a VCR with Flying Eraser Head, and so forth. In many cases, such new features are heavily advertised and emphasized when they are first introduced in the market.

However, while some of those features are actually used by consumers, others are very rarely used. For instance, Steady Shooting feature of a camcorder is a built-in function; the feature is used once the power is turned on. The Woofer Sound feature may or may not be used depending on the user’s preference. On the other hand, features such as Flying Eraser Head of a VCRCthat often require user skillsCare rarely used and may never be used in the future, despite their usefulness . Thus, from a consumer’s standpoint, these product features may be useless and irrelevant to purchase decision. Then one interesting and important question arises: Why do some people spend money on the product features that they may never use?

If consumers are sure that they will not use certain features, the rational purchase decision would be not to buy the product with such features, unless the product with the feature provides a lower price. However, this may not always be the case in reality. For some consumers, the utility of possessing the "top of the line" model may override any other disutilities such as paying more dollars. For others, there may be a tendency to overestimate the probability of using those features in the future. In general, how do those additional features influence an individual consumer’s choice set? How do they affect the consumer’s purchase decision? What types of judgmental processes underlie such decisions? These questions inspired this study. However, it is not the intent or the scope of this paper to answer all these questions. The purpose of this paper is twofold: (a) to demonstrate that the tendency really exists for some consumers to buy features that they may not use, and (b) to determine whether purchase decision situations and brand characteristics influence such a tendency.

Initial Considerations. First of all, this study would need a limiting condition for the product features that are in question. The first boundary condition for the product features being studied in this study is the ambiguity or uncertainty surrounding the usage and/or value of the feature or sometimes the product itself. This is not an unreasonable assumption in the high-tech marketplace.

The second limiting condition for the features examined in this study is that they are largely inseparable from the focal/main product and do not carry any independent value when they are separated from the product (such as the Steady Shooting for camcorders). As such, the additional features in this study do not include product bundles such as travel packages which often include airfare, hotel reservation, rental car and so forth. This condition implies that only the manufacturers control the distribution or incorporation of the features; consumersConce bought the featureCcannot even sell them in a secondary market, for example. Manufacturers typically provide several optional models; some with feature A, some with feature B, some with both A and B, still others without any new features. Consequently, a consumer has to select from the set of models as predetermined by manufacturers. The only choice a consumer can make regarding this feature is either to buy the entire camcorder that includes the desired feature or to buy another camcorder without the feature, once he or she is to buy a camcorder. Then it follows that, if there exists only one product model in the marketplace and that model happens to have one or more features with doubtful perceived value, consumers may have no choice but to buy the whole product anyway. Thus, an important assumption in this study is that consumers have at least two alternative models to consider, that is, one with a new feature and one without the feature.


Nowlis and Simonson (1996) investigate factors that moderate the impact of a new feature on brand choice. In addition to the characteristics of the feature itself, using two principlesCmultiattribute diminishing sensitivity and performance uncertaintyCthey propose that the characteristics of the product to which new features are added are important determinants of the impact of these features on sales and market share. Specifically, Nowlis and Simonson argue that when a new feature is added to a brand with superior features or brand name, the assimilation effect would occur (Lynch, Chakravarti, and Mitra 1991; Martin, Seta, and Crelia 1990). Because of the existing perception of superior performance, the new feature is not expected to significantly increase the overall evaluation of the product. On the contrary, when a new, unique feature is added to an inferior alternative or weaker brand, the contrast effect would occur. Thus, the new feature is likely to have a relatively large positive impact on the overall product evaluation (see also Srull and Wyer 1989). However, Nowlis and Simonson did not examine the underlying reasons why consumers might prefer the products with new features with an uncertain future value/usage.

As was evidenced by Norman (1988), many consumers find high-tech equipment such as computers, video cameras, and VCRs too complex and difficult to use. Apparently, most people do not have the appropriate procedural knowledge to use the timed recording feature on their VCRs. Thus the lack of user skills or procedural knowledge can be a good reason for not using the product features that consumers have bought already. Do the consumers, then, overestimate their technical capabilities or user skills at the time of purchase?

Another related issue is the potential perceptual uniqueness of the additional features. According to Carpenter, Glazier, and Nakamoto (1994), companies can make meaningful differentiation from a meaningless attribute, that is, from a unique but irrelevant attribute. For example, they added irrelevant product attributes (such as "Alpine class" fill in down jackets, "authentic Milanese" style pasta, and "studio-designed" signal system in compact disc players). What if we consider the same problem at the level of product features? Again, for many consumers certain features are "irrelevant" because the features will not be used in the future. Following the logic of Carpenter and his colleagues, the additional features may increase the preference for the product becauseCjust as attributes did in their studyCthe features can provide uniqueness or newness. Once the additional features are perceived as unique, then it will affect the perception of overall evaluation of the product, even if the additional features are unimportant or irrelevant to the consumer. This in turn can provide a reason why people spend money on the feature they will not use. Likewise, it seems that there could be many alternative easons for people’s overspending on the new product features.


This paper proposes and explores two mechanisms as the possible key explanations for the hypothesized phenomenon, that is, why people spend money on the feature that they will not use: (1) expected future value and (2) risk aversion. As discussed in the preceding section, there could be many reasons for the consumer’s overspending on new features, weCin this paperCdecided to focus on the foregoing two reasons in this paper. We will further explore these two possibilities in the following.

Preference Reversal?

Economists argue that consumers have preference for short-term gains even in the face of potentially higher long-term costs (Abelson and Levi 1985). In other words, people often discount the future values; e.g., a dollar today is worth more than a dollar a year from now (inflation aside). Similarly, people often take future costs and benefits less seriously than those of present times. However, such a preference pattern appears to be "reversed" when consumers are buying new product features that they are not familiar with. This is likely to happen when the consumer overestimates either the future value or the future risk associated with the product feature.

Consumers may be overreacting to the possibility of future purchase failure if they did not buy the new feature. The extra money spent on the additional feature is a sure loss consumers generally want to avoid. However, what if the current "savings" on that feature make the total spending on that product a mistake? If the consumer believes that the additional feature the consumer did not purchase can become a standard component (e.g., car seat belt, or Pentium processor) in the near future, the present purchase may turn out to be a complete purchase failure which may necessitate the purchase of the entire product including the new feature. The potential loss in this case, for sure, would be greater than the cost of buying an additional feature at present time. This implies that the environmental changes and the future purchases can affect the present buying decisions.

One of theoretical backgrounds of the above argument can be anticipated regret (Janis and Mann 1977). When mentally rehearsing or examining the question "Will I regret the current decision later?" anticipated regret appears to be a counteracting mechanism against discounting the future. Thus, anticipated regret may result in reversing a short-run decision. In this study, the possible purchase failure is the source of regret. Put together, most consumers are risk averse, and one way of avoiding risk in this study may be 'just buying a product with the additional feature.’ This risk aversion, in turn, will demonstrate the tendency of the hypothesized phenomenon.

Another way to explain the purchase of additional features is to think of the purchase as investment. In this case, consumer may not have explicit expectation about return from the investment. Consumer expectation could be based on a really simple heuristic: "I might use this feature at a future point in time." Nonetheless, buying additional features can definitely be an investment if the consumer has some expectations. The issue is : "Can such a simple expectation lead the consumer to buy the product with the feature?" If this is the case, then future expectation (that is, I might use this new feature in the future) provides an explanation for the hypothesized phenomenon. This future expectation may or may not bias the probability of using the feature depending on individual characteristics such as prior knowledge about the product, familiarity, and so forth. However, if consumers overestimate the probabilty , then the hypothesized phenomenon is likely to occur. The above arguments lead to the following hypotheses.

H1: Consumers who are concerned about future expectation are more likely to buy additional features than those who are not.

H2: Consumers who are sensitive to purchase risk (a possible future purchase failure) are more likely to buy additional features than those who are not.

Effects of Brand Reputation

Depending on brand, the tendency of consumers to buy products with additional features may differ; specifically, between less-known and well-known brands. Well-known brand names, compared to less-known ones, would serve as a halo and elicit more favorable evaluation from consumers for the additional features. In purchase situations involving risk aversion and performance uncertainty, brand names can be used as an indicator of the product quality. They can also affect consumer perceptions of the potential usage of the additional features.

H3: Consumers are more likely to buy additional product features of well-known brands than that of less-known ones.

Even though purchase intentions, once framed with future expectations (FE) or risk aversion (RA), would be greater than the conditions without decision framing for both less-known and well-known brands, the impact of future expectation and risk aversion decision framing would vary depending on brand reputations. For well-known brands, for example, future expectation framing condition would not have a far bigger impact on purchase intention than the condition without future expectation framing because of the existing perception of the well-known brand. On the contrary, for less-known brands, purchase intention would be significantly increased with decision framing (FE or RA) because consumers would want to avoid purchase failures from less-known brands. In other words, a brand-by-decision frame interaction is expected.

H4: The impact of decision framing will vary depending on brand. Specifically, the impact of decision framing (either FE or RA) will be bigger with less-known brands than with well-known brands.


The basic premise underlying this paper is that some people purchase products with features they may never use. Study 1 was conducted to determine if such a hypothesized phenomenon really occurs in the marketplace. A survey was conducted to achieve the goal.


Three different types of productswith various price ranges were selected: TV set, VCR, and cassette tape player. Additional features were determined and their functional descriptions were provided in the survey. Those features were Picture-in-picture (PIP) function for the TV set, Flying Eraser Head for the VCR, and Recording function for the cassette player. Sixty-five respondents completed the survey in which they were asked about their experiences with the products and the features in terms of purchase and usage. Specifically, they were asked whether or not they had bought the products, and, if they had, how often they had used the specified features. In addition, they were asked their potential usage levels in the future. Another question was how important the features were when they bought the above products.


As expected, the results of the survey showed extremely low usage levels for the three features. Even for those who had the products and had used the features, the usage levels for the three features were very low. The average number of times they had used the features over the past three months was only 5.3. For instance, out of 65 respondents, only 18 had ever used the Picture-in-picture feature of their TV set and they had used it, on average, only 3.2 times in the last three months. For the VCR, only three respondents reported to have used the Flying Eraser Head function, and the average usage frequency was only 1 for the past three months. Compared to the other two features in the survey, the Recording function of the cassette player is easier to use; the only required operation is just to push the recording button. Nevertheless, the usage level for this feature was also extremely low.

Respondents were also asked to give an approximate, potential number of usage for the next three months. The average potential usage rates were 6.7, 1.4, and 1.9 times for Picture-in-picture, Flying Eraser Head, and Recording functions, respectively. The Picture-in-picture function had the highest average because of one person who responded that he would use it everyday (90 times for three months). However, the modes of the potential usage rates for all three features were zero and the medians were less than one for all three. The importance of these features to purchase decisions was rated; the mean values were 2.5, 2.1, and 2.5 respectively on a 7-point scale (1= "not important at all"; 7="very important"). This also implies that the above features were not perceived important when consumers made buying decisions. The results of the survey, taken together, suggest that the hypothesized phenomenon does occur among consumers.


Study 2 focused on testing the hypotheses in this study. The product feature used here is Flying Eraser Head feature for VCRs. This feature was selected because its characteristics were relatively less-known to general population. It was an important feature performing a special task, but not important enough to affect purchase decisions of VCR’s, as was shown in Study 1. Nonetheless, if buyers would show a tendency of adopting the product with this feature, then much stronger conclusions could be drawn about the hypothesized phenomenon.


The first factor examined in this study was whether the brand was well-known or less-known. Sony was used as a well-known brand whereas a fictitious brand, S&T, was used as a less-known brand. Subjects were told that S&T was a new VCR manufacturer in California. The second factor was "future expectation" (FE) decision frame that tried to enhance the level of future expectation by saying that the additional feature would have a huge potential for future purposes. The third factor was "risk aversion" (RA) tha tried to induce risk aversion tendency by providing a scenario saying that many people regretted not having bought the additional feature. In sum, this study used a 2 (brand: well-known vs. less-known) X 2 (FE: Yes vs. No) X 2 (RA: Yes vs. No) between-subject design.

The dependent variable for this study was the purchase intention of the VCR with the Flying Eraser Head, rated on a 7-point scale where 1="definitely will not buy" and 7="will definitely buy."

Subjects and Procedure

Data were collected at a church in a Mid-Western town. A total of 160 subjects participated in this study. Subjects were given a questionnaire with a scenario describing a decision making situation. They were randomly assigned to one of the eight experimental conditions. After given a brief introduction of the study purpose, they were asked to read and envision the scenarios in the questionnaire as if they were in the situations. All subjects received the same information about the functional characteristics of the Flying Eraser Head feature.

For the subjects who received both the FE and RA frames, the order of FE and RA was counterbalanced.


Manipulation Check. Subjects who expressed their interest in purchasing a VCR with the Flying Eraser Head feature were asked to choose from the following list the two reasons why they would buy one: (a) "Flying Eraser Head is an important feature in general," (b) "I might use the feature in the near future," (c) "I might regret in the future if I don’t buy the feature," (d) "The price difference is not big between the two types of models and," and (e) "It’s a Sony (or S&T) product." The frequency distribution of the subjects’ responses to this question served as a check fo the decision framing manipulation. In the FE condition, 12 subjects out of 21 selected (b) whereas in the RA condition, 10 subjects out of 23 selected (c). In the FE-RA condition, 17 subjects out of 29 selected either (b) or (c).

Hypotheses Testing. The results showed that there was a significant main effect for both decision frames (p<.000). The mean values of the purchase intention for those who received a FE scenario and those who did not were 5.11 and 4.06 respectively on a 7-point scale. For the RA frame, the mean values for those who received a RA scenario and those who did not were 5.06 and 4.11 respectively. Thus, H1 and H2 were supported (see Table 1).

The mean ratings of the purchase intention were not significantly different between the well-known brands and the less-known brands. The means of less-known and well-known brand were 4.45 and 4.73 respectively. Although the mean value for the well-known brand is a little large, the main effect of brand was not significant (p=.287). Thus, H3 was not supported. However, the results showed that there was a (marginally) significant interaction effect between brand and both of the decision frames. For the less-known brand, the mean values of the condition with FE and without FE were 5.23 and 3.68 respectively; for the well-known brand, 5.00 and 4.45, respectively. This result implies that with FE condition, brand does not influence much the purchase intention. However, without FE condition, brand may provide an explanation for the greater purchase intention, that is well-known brands are preferred. The brand-RA interaction also showed a similar result. For the less-known brand, the mean values of the condition with RA and without RA were 5.18 and 3.73 respectively; for the well-known brand, 4.95 and 4.50, respectively. Therefore, H4 was supported. Again, this implies that brand may not be a significant factor when people are framed to have risk aversion about additional features. Considering the quality uncertainty of less-known brands, from this interaction effect, one can infer that consumers might feel tat the safest way to avoid risk is just to buy the additional feature.




Study 1 shows the hypothesized phenomenon does exist; many consumers do not use some features after they have bought the product. Study 2 examines the conditions under which consumers buy the product with those features they may never use. It demonstrates that consumers show different levels of purchase intentions depending on how the decision situation is framed. It also shows that brand name influences the purchase intention under certain decision situations. However, there seem to be some alternative explanations about the phenomenon in the study. Before discussing other explanations, the limitations of the study should be mentioned.

Regarding the measurement issues, asking people to recall what was important to them in the past when currently it is not important to them may bias the response in Study 1. There exists a possibility that the described features were, in fact, very important to respondents at the time they made purchases. The manipulation check for Study 2 was not quite precise in that it did not adopt a continuous scale. Regarding the brand manipulation, there could be confounding effect of country of origin. We could have more solid results if we used the "new" brand from the same country as the well-known brand, that is, S&T from Japan rather than from California. Keeping this constant could prevent other inferences from adding noise to the data.

From a theoretical perspective, FE and RA factors may not be truly orthogonal. These two factors can create confusion. Because FE is operationalized as "you will definitely need this new feature" and the RA as "purchases without new features are a failure because buyers now need ," (see Appendix 1) both can be interpreted as representing "need." In addition, the magnitude of two effects should be addressed. Depending on the framing of the buying situation, the choice (purchase the product with the additional feature) can be viewed as either an investment or a risk aversion. These two behaviors can be regarded as gain or loss. Risk aversion from regret is loss from the prospect theory’s perspective, which has a convex function. Investment is viewed as gain, which has a concave function. According to prospect theory (Kahneman and Tversky 1979), a loss function has a steeper slope than a gain function. This implies anticipated regret may be a stronger explanation than investment. Thus, the effect of risk aversion is expected to be larger than that of future expectation, even though both are expected to increase the probability of buying additional features. This argument can be a subject of future studies.


This study shows that there exists an overestimation of probability of using additional features in the future depending on brand name and decision framing. However, still other possible explanations are waiting for their verifications.

Decoy Effect. Previous research demonstrates that adding a decoy to a choice set can increase choice shares of brands similar to the decoy while reducing shares of brands dissimilar to it. Heath and Chatterjee (1995) suggest that a decoy reduces shares of lower-quality competitors more than it reduces shares of higher-quality ones. Then what happens if a superior alternative instead of a decoy is added to a choice set? Tversky and Shafir (1992) examine the effect of added alternatives using s superior CD player in one case and an inferior one in another. They show that the former case creates high conflict in a decision maker’s mind because s/he needs to determine whether s/he is better off with a cheaper and popular product, or with a more expensive and sohisticated one. Since this conflict is not easy to resolve, some people delay their purchase until they learn more about various alternatives. In this case, the presence of an inferior alternative tends to increase the market share of a dominating one (see also Huber, Payne, and Puto 1982).

This line of logic, applied in the present research context, suggests that the product with an additional feature may be viewed as an improved , high quality brand while the existing ones without it may be perceived as if it is a decoy. Once they are perceived as decoys, the market share of the superior alternative (the one with an added feature) will increase (Huber et al. 1982). In other words, the product with an added feature is more likely to be adopted by consumers. This judgmental process may be an alternative explanation about why people purchase products with features they may not use.



Sunk Cost Effect. People often make a systematic error of continuing to buy or invest on something because of sunk cost. The sunk cost consideration comes into play once an investment in money, effort, or time has been made (Arkes and Blumer 1985). When people have already made a commitment to buy a certain product (e.g., a computer), they would buy an additional feature (e.g., a fax modem) even when the feature has little value to them. In other words, they evaluate the product not in terms of the incremental price for the additional feature but in terms of the total price of the product including sunk costs. Thus, those who consider sunk costs in their purchase decision may overestimate the usefulness of the additional feature, compared to those who do not. Such a tendency becomes evident especially when the price difference is not great between the model with the additional feature and the one without it.

Price Effect. Another issue is the effect of price. If the price of products is high, then the anticipated regret will be greater for not buying additional features, because the magnitude of perceived risk of purchase failure will be greater than that of low-priced products. For low-priced products, people may perceive buying additional features as an investment, because the impact of purchase failure (if the purchase decision turns out to be a complete failure) would not be as great as those of high-priced items. For example, people may think spending three dollars more is no big deal when they buy a twenty dollar product. Thus, for high-priced products, anticipated regret will play a key role, and for low-priced products, investment will play a role in buying additional features.


Abelson, Robert P., and Ariel Levi (1985), "Decision Making and Decision Theory," In G. Lindzey and E. Aronson (Eds.), The Handbook of Social Psychology (3rd Ed.), Vol.1. 231-309.

Arkes, Hal R. and Catherine Blumer (1985), "The Psychology of Sunk Cost," Organizational Behavior and Human Decision Process, 35, 124-140.

Carpenter, Gregory S and Rashi Glazer, and Kent Nakamoto (1994), "Meaningful Brands From Meaningless Differentiation: The Dependence on Irrelevant Attributes," Journal of Marketing Research, 31(August), 339-350.

Heath, Timothy B. and Subimal Chatterjee (1995), "Asymmetric Decoy Effects on Lower-Quality versus Higher-Quality Brands: Meta-analytic and Experimental Evidence," Journal of Consumer Research, 22 (December), 268-284.

Huber, Joel, John W. Payne, and Christopher Puto (1982), "Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis," Journal of Consumer Rsearch, 9 (June), 90-98.

Janis, I.L., and L. Mann (1977), Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment, New York: Free Press.

Kahneman, Daniel and Amos Tversky (1979), "Prospect Theory : An Analysis of Decision Under Risk," Econometrica, 47, 263-291.

Lynch, John G., Dipankar Chakravarti, and Anusree Mitra (1991), " Contrast Effects in Consumer Judgments: Changes in Mental Representations or in the Anchoring of Rating Scales?" Journal of Consumer Research, 18(3), 284-297.

Martin, Leonard L., John J. Seta, and Rick A. Crelia (1990), "Assimilation and Contrast as a Function of People’s Willingness and Ability to Expend Effort in Forming an Impression," Journal of Personality and Social Psychology, 59 (1), 27-37.

Norman, Donald A. (1988), The Psychology of Everyday Things, New York: Basic Books.

Nowlis, Stephen M. and Itamar Simonson (1996), "The Effect of New Product Features on Brand Choice," Journal of Marketing Research, 33 (February), 36-46.

Srull, Thomas K. and Robert S. Wyer, Jr. (1989), "Person Memory and Judgment," Psychological Review, 96 (1), 58-83.

Tversky, Amos, and Eldar Shafir (1992),"Choice Under Conflict: The Dynamics of Deferred Decision," Psychological Science, 3, 6, 358-361.



Jaebeom Suh, University of Alabama at Tuscaloosa
Yong-Soon Kang, State University of New York at Binghamton
Moonkyu Lee, Yonsei University


NA - Advances in Consumer Research Volume 25 | 1998

Share Proceeding

Featured papers

See More


E4. Doing Good for Nothing: Motive Inferences from the Probabilistic Profits of Prosociality

Ike Silver, University of Pennsylvania, USA
Jackie Silverman, University of Pennsylvania, USA

Read More


Anchors as Midpoints: it’s not the Size of the Adjustment that Counts, it’s the Direction

Joshua Lewis, University of Pennsylvania, USA
Joseph P. Simmons, University of Pennsylvania, USA

Read More


Deny the Voice Inside: Are Accessible Attitudes Always Beneficial?

Aaron Jeffrey Barnes, University of Illinois at Urbana-Champaign, USA
Sharon Shavitt, University of Illinois at Urbana-Champaign, USA

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.